Two major barriers exist within the tumor bearing host that prevent the efficient generation of tumor specific effector CD8 cells. First, as many tumor antigens are also expressed on healthy tissue, these antigens have already been presented to the immune system, as toleragens. As a result, the most efficacious high affinity T cells may have already been eliminated through self-tolerance. Second, the tumor specific T cells that can be stimulated are difficult to maintain as effector cells, and perhaps, even more difficult to maintain as memory cells. This proposal will utilize a model system to study tumor immunity in which a surrogate tumor antigen, the influenza hemaagglutinin (HA), is expressed as a self-antigen on healthy tissue in the pancreas of InsHA transgenic mice. The goals of this proposal are to determine how expression of tumor antigen by normal tissue affects tumor immunity and to develop strategies that can override this tolerance barrier at the stimulatory, effector and memory stage of CD8 development. In order to achieve these goals, they have developed a TCR transgenic line expressing low affinity HA specific TCR derived from an InsHA mouse. This, together with the high affinity HA specific Clone 4 TCR transgenics, will allow us to directly monitor the fate of naive, activated and memory T cells specific for HA under a variety of experimental manipulations designed to enhance the activation and longevity of effector cells in tumor bearing mice. Techniques to be used for immunization of tumor specific CD8 cells include virus that express the HA antigen, DNA immunization, and peptide immunization assisted by anti-CD40MAb. Strategies that will be tested to increase longevity and efficacy of activated effectors include 1. Antibody-cytokine fusion proteins that will direct to the tumor key cytokines, such as IL-2 and GM-CSF, that may enhance activation and longevity of effectors. 2. Anti-CD40 MAb that may enhance APC function required for stimulation of tumor specific T cells, and 3. CD30-Ig, which has been shown to -prevent apoptosis of effector T cells. Another goal of this proposal is to determine if it is possible to maintain memory T cells specific for a tumor antigen (that is shared by normal tissue) at a level required to prevent the growth of tumor cells. The hypothesis will be tested that tumor specific memory T cells may be subject to the same types of peripheral tolerance mechanisms that result in elimination of naive T cells.